In-vitro testing of RF–enabled low force mechanical thrombectomy

for ischemic stroke

Chi Hang Chon, Zhen Qin, Alexander KN Lam, John CK Kwok, Matthew MF Yuen and David CC
Lam

thrombolysis has a 3 to 6 hours treatment time window, less



Abstract— Mechanical thrombectomy for ischemic stroke

has high recanalization rate, long treatment time window and than 50% recanalization rate and high risk of symptomatic
low hemorrhage risk. However, the clot engagement approach intracerebral hemorrhage (sICH) [4-6]. Because of these
of caging the clot against the vessel wall can cause vessel limitations, only approximately 5% of patients are eligible for
stenosis and stroke recurrence. A device with reduced radial the treatment [7].
stenting force that reduces vessel wall friction would minimize
stenosis and damage. The use of localized Radio Frequency Unlike chemical thrombolysis where the drug courses
(RF) to enable clot engagement and retrieval with minimal throughout the entire circulatory system of the body,
stenting force is explored in this study. New mechanical mechanical thrombectomy is effected only locally at the
thrombectomy devices enabled with RF (Patent No.: US occlusion. Distal device, proximal device and stent retriever
62/172,043) were built and tested on human blood clots in are three major types of mechanical thrombectomy devices;
vessels ex vivo. Test results showed that the RF-mechanical and stent retriever is considered as amongst the more robust
thrombectomy successfully and reproducibly captured and type of device [9].
retrieved the clots without relying on stent caging of the clot
against the vessel wall. Further work will be conducted on In minimally invasive surgical procedure, stent retriever
animals to compare vessel wall damage between conventional is threaded endoluminally to the occluding thrombus. With
and RF-mechanical thrombectomy. successful retrieval of the clot, the blood flow is restored in
under half an hour [9]. The procedure has a recanalization
I. INTRODUCTION rate up to 82% [6, 8], but the shortcoming is that local vessel
damage may result from the engagement and retrieval
Stroke is the second leading cause of death with 0.61 process.
million new cases in the U.S. [1] and 2.5 million new cases in
China [2], and 6.7 million deaths per year worldwide [3]. In Vessel damage in mechanical thrombectomy results from
Caucasian population, 87% of cases are ischemic stroke the manner in which the device cages the clot and interacts
while the remaining are hemorrhagic strokes [1]. Ischemic with the vessel wall. In mechanical thrombectomy, the
stroke results from occlusion of cerebral arteries by a device expands radially and cages the clot by forcing the clot
thrombus or an embolus that interrupts supply of blood to the against the vessel wall. Poor clot engagement and
brain. Prolonged interruption will result in brain damage, incomplete retrieval may result if the expansion force is
disability and possible death. insufficient [9]. High radial stenting force will increase
engagement security, but post-operation examinations show
Current treatments for ischemic stroke include chemical that vessel damage may result from the use of a large stenting
thrombolysis and mechanical thrombectomy. Thrombolytic force [10-13].
drugs, such as recombinant tissue plasminogen activator
(rtPA), are injected intra-venously or intra-arterially to Histological analysis on vascular damages in swine model
chemically break down the clot. Typical chemical by B. Gory, et al [10] had shown that mechanical
thrombectomy devices can cause vascular damage, such as
endothelial denudation and intimal edema, and vessel
C. H. Chon is with the Department of Mechanical and Aerospace stenosis and stroke recurrence which worsen the clinical
Engineering, The Hong Kong University of Science and Technology,
Kowloon, HK (Phone: +852-2358-8661; e-mail: [email protected]).
outcomes may result.
Z. Qin is with the Department of Mechanical and Aerospace A better clot engagement method is needed to reduce
Engineering, The Hong Kong University of Science and Technology,
Kowloon, HK (e-mail: [email protected]). vessel stenosis and damage. In this study, a new patented
A. K. N. Lam is with the Department of Mechanical and Aerospace RF-mechanical thrombectomy device has been developed.
Engineering, The Hong Kong University of Science and Technology, Experiments were conducted in this study to examine the
Kowloon, HK (e-mail: [email protected]). engagement and retrieval behavior of the new device.
J. C. K. Kwok is with Division of Biomedical Engineering, The Hong
Kong University of Science and Technology, Kowloon, HK and with the
Department of Neurosurgery, Kwong Wah Hospital, Kowloon, HK (e-mail: II. MATERIAL AND METHOD
[email protected])
M. M. F. Yuen is with the Department of Mechanical and Aerospace A. RF-Mechanical Thrombectomy
Engineering, The Hong Kong University of Science and Technology, Mechanical frictional caging used in mechanical
Kowloon, HK (e-mail: [email protected]). thrombectomy can be replaced if the device can adhesively
D. C. C. Lam is with the Department of Mechanical and Aerospace
Engineering, The Hong Kong University of Science and Technology,
engage the clot. Blood clots contain proteins such as
Kowloon, HK (e-mail: [email protected]). albumins, globulins and fibrinogens. These proteins adsorb
naturally on metallic surfaces. Interfacial adhesion between
978-1-4244-9270-1/15/$31.00 ©2015 IEEE 1349
the clot protein and the surface is protein-surface adsorption After positioning the clot, a syringe was then connected
interaction. Protein adsorption can be enhanced by using an to the distal end of the Tygon tube to maintain the pressure
energetically favorable surface, changing the pH or within the tube.
increasing the temperature [14, 15]. Electrical current with
frequency higher than 100 kHz applied onto a wire surface is
known to enhance protein absorption and adhesion on
exposed wire surfaces without causing side effects.
Clinically, RF devices with frequency between 200 kHz and
5 MHz have been used in surgery. A principal feature of
these devices is their ability to quickly induce clotting. RF
energy is focused onto the point of application and the Figure 1. Prototype of the new RF-mechanical thrombectomy device
clotting is instantaneous. Instead of focusing the RF energy
to a point, a large clot can be engaged using zonal RF on a
RF-mechanical thrombectomy device. In this study, we will
examine the effect of zonal RF on the adhesion strength of E. RF Thrombectomy Procedure for in-vitro test
the clot on metallic surfaces The RF thrombectomy procedure was a three-step
B. Prototyping of RF-mechanical thrombectomy device procedure: navigation and deployment of device, application
of the RF to induce adhesive clot engagement, and retrieval
A RF zone can be created between two or more wires. of the clot and device. The RF thrombectomy device was
Prototypes of RF-mechanical thrombectomy were fabricated initially stowed inside a catheter and navigated through the
from superelastic nitinol wire (0.25 mm dia) and stainless hemostatic valve, along the y-connector and into the Tygon
steel wire (0.125 mm dia). Superelastic nitinol wire was used tube. After the catheter penetrated the clot (Fig. 2), the
as the backbone of the device prototype, and device was deployed by unsheathing the catheter while
polytetrafluoroethylene (Teflon) tubes of various diameters holding the device in position so that the clot was positioned
were used as insulators (Fig. 1). in the middle of the RF zone. After positioning the device, a
By shaping the stainless steel wire into a loop and RF was activated for a controlled period of time. After
attaching the wires to a polytetrafluoroethylene (Teflon) tube successful adhesive engagement, the clot was then retrieved.
as shown in Fig. 1, clot adhesion could be promoted in the
zone between the two wires. To avoid a large stenting force
against the vessel wall, the device was fabricated to be
within 4 mm, the inner diameter of the test vessel.
RF was generated by connecting the proximal ends of the
nitinol wire and stainless steel wire prototype connected to a
200 kHz RF system operating at 10V, but would be step up
to higher voltages when in use. Figure 2. Thrombectomy procedure for in vitro experiment

C. Clot Preparation
Human whole blood was obtained from healthy donors.
The blood was stored in citrated blood bags in a refrigerator
at 4 oC prior to use. To prepare clot samples for tests, 6 ml
of citrated blood was mixed with 1 ml of calcium gluconate
solution. The blood was then injected into a Tygon tube
section (4.76 mm inner diameter) and stored in a chamber at
37 oC for 30 minutes for coagulation. After 30 minutes, the Figure 3. Clot in initial position of experiment set-up. Inner tube was
lubricated to mimic wet endothelium surface before use.
Tygon tube section was removed and the clot was then
carefully cut into samples of 5 mm in length using a scalpel.
D. Occlusion Model
A 3.17 mm inner diameter Tygon tube was formed into a
curved shape with 20 mm radius of curvature. A Y-
connector was connected to the proximal end of the Tygon
tube. A clot sample was placed in the distal end of the Tygon
tube and suction was applied using a 5 ml syringe with the
hemostasis valve being closed as shown in Fig. 3. Suction Figure 4. Clot retrieved to target position
was applied until the clot was positioned at the designated
location in the tube. 10 trials were performed for every RF pulse duration (0s,
1s, 3s or 5s). Engagement was deemed successful if the clot
was retrieved to the proximal end of the Tygon tube as
shown in Fig. 3 and Fig. 4. Weights of the clot and device

1350
were measured before and after the thrombectomy procedure showed that only liquid remained in the tube after RF
to quantify the completeness of the retrieval. duration. This suggests that liquid was squeezed out from
the clot after RF duration.
F. Temperature Measurement
When RF was applied, maximum temperature would be TABLE 1. WEIGHT OF CLOT SAMPLE AND RETRIEVED CLOT
induced in the clot between the metallic wires. To determine RF Weight of Weight of Percent Remarks
the clot surface temperature distribution under different RF Duration clot sample retrieved retrieved
treatment, a RF thrombectomy device was embedded in a and Device clot and
(g) Device (g)
semi-cylindrical clot (Fig. 5), so that the device would be
close to the top surface of clot. The flat clot surface 0s 0.59, 0.60, 0.47, 0.53, 8%, 50%, Solid
temperature was characterized using an infra-red camera 0.61 0.47 0% residues
(Ti25, Fluke) for clot treated with RF for durations ranging 3s 0.54, 0.58, 0.50, 0.51, 45%, 50%, Liquid
from 3s, 5s to 10s. The temperature was measured in 2- 0.57, 0.58, 0.52, 0.51, 42%, 50%, residues
second interval over a 60s period. 0.58 0.52 36%
5s 0.54, 0.54, 0.50, 0.50, 56%, 50%, Liquid
0.55, 0.59, 0.51, 0.51, 56%, 33%, residues
0.57 0.51 45%

Figure 5 Experimental setup for temperature measurement and geometry

of the mold

III. RESULTS
A successfully engaged and retrieved clot with the device
is shown in Fig. 6. The whole clot was retrieved to the
proximal end of the Tygon tube without fragmentation.
Figure 7. Engagement success rate

Figure 6. Clot retrieval procedures

Preliminary engagement success statistics as a function

of RF duration are plotted in Fig. 7. 25% successful retrieval
rate was obtained in the control group (0s); 80% successful
retrieval rate was obtained using 1s RF duration; and 100%
Figure 8. Weight of clot sample and retrieved clot
retrieval rate was obtained with 3s and 5s RF durations.
Weight of clot and device before and after retrieval using Temperature profile as a function of time is plotted in Fig.
0s, 3s and 5s RF durations are shown in TABLE 1. The 9. When RF was applied, temperature of clot increased
average weights are plotted against 0s, 3s and 5s RF rapidly and reached a peak value. Temperature then
durations in Fig. 8. The data showed that 81% of the clot decreased gradually. The peaks temperature for the 3s , 5s
mass remained in the control group (0s); 55% of clot mass and 10s treatment were respectively of 45 oC, 57 oC and 58oC
remained after 3s RF duration; 52% of clot mass remained respectively.
after 5s RF duration. Observations of the emptied tube
1351
 force that may damage the vessel wall and lead to stenosis.
Further studies in animals will be conducted to determine the
efficacy of the procedure in vivo.

ACKNOWLEDGMENT
We would like to thank the staff of the Kwong Wah
Hospital for their assistance.

REFERENCES
[1] A. S. Go, D. Mozaffarian, V. L. Roger, E. J. Benjamin, J. D. Berry,
M. J. Blaha, S. Dai, E. S. Ford, C. S. Fox, S. Franco, H. J. Fullerton,
C. Gillespie, S. M. Hailpern, J. A. Heit, V. J. Howard, M. D.
Huffman, S. E. Judd, B. M. Kissela, S. J. Kittner, D. T. Lackland, J.
H. Lichtman, L. D. Lisabeth, R. H. Mackey, D. J. Magid, G. M.
Marcus, A. Marelli, D. B. Matchar, D. K. McGuire, E. R. Mohler 3rd,
C. S. Moy, M. E. Mussolino, R. W. Neumar, G. Nichol, D. K. Pandey,
Figure 9 Temperature profile during RF application N. P. Paynter, M. J. Reeves, P. D. Sorlie, J. Stein, A. Towfighi, T. N.
Turan, S. S. Virani, N. D. Wong, D. Woo, M. B. Turner and American
Heart Association Statistics Committee and Stroke Statistics
IV. DISCUSSION Subcommittee, "Heart disease and stroke statistics--2014 update: a
report from the American Heart Association," Circulation, vol. 129,
The clot adhered to the wire after RF duration of 3
pp. e28-e292, Jan 21, 2014.
seconds or longer. RF enhanced adhesion and enabled clot [2] L. Liu, D. Wang, K. S. Wong and Y. Wang, "Stroke and stroke care in
retrieval without the need for a large clot engaging stenting China: huge burden, significant workload, and a national priority,"
force used in conventional mechanical thrombectomy. Stroke, vol. 42, pp. 3651-3654, Dec, 2011.
[3] A. Alwan, Global Status Report on Noncommunicable Diseases 2010.
Weight loss analysis revealed that the clot lost weight after World Health Organization, 2011.
RF. Observation showed that only liquid remained in the [4] W. Burnette and G. Nesbit, "Intra-arterial thrombolysis for acute
test vessel. This suggests that liquid had been squeezed out ischemic stroke," Eur. Radiol., vol. 11, pp. 626-634, 2001.
[5] A. Ciccone, L. Valvassori, M. Nichelatti, A. Sgoifo, M. Ponzio, R.
from the initial clot, making the clot denser and stronger. Sterzi and E. Boccardi, "Endovascular treatment for acute ischemic
The denser clot is an indication that the clot had been stroke," N. Engl. J. Med., vol. 368, pp. 904-913, 2013.
strengthened by the RF. The strength of the clot as a [6] P. Mordasini, C. Zubler, G. Schroth and J. Gralla, "Thrombectomy for
acute ischemic stroke treatment: a review," EJMINT Invited Review,
function of RF application will be examined in a future vol. 1238000077, 2012.
study. Temperature data showed that the temperature [7] T. Liebig, J. Reinartz, R. Hannes, E. Miloslavski and H. Henkes,
rapidly reached the peak and decreased gradually. Studies in "Comparative in vitro study of five mechanical embolectomy systems:
effectiveness of clot removal and risk of distal embolization,"
the literature showed that damage is a function of both Neuroradiology, vol. 50, pp. 43-52, 2008.
temperature and duration. The duration of peak temperature [8] G. Gascou, K. Lobotesis, P. Machi, I. Maldonado, J. F. Vendrell, C.
was short and the peak temperature was not far from the Riquelme, O. Eker, G. Mercier, I. Mourand, C. Arquizan, A. Bonafe
and V. Costalat, "Stent retrievers in acute ischemic stroke:
body temperature. This suggests that thermally induced
complications and failures during the perioperative period," AJNR
damage is minimal based on the time-temperature analysis Am. J. Neuroradiol., vol. 35, pp. 734-740, Apr, 2014.
by R. Agah, et al. [16]. However, actual effects on live tissue [9] N. Beadell and H. Lutsep, "New stent retriever devices," Curr.
should be further investigated in ex-vivo and in-vivo Atheroscler. Rep., vol. 15, pp. 1-6, 2013.
[10] B. Gory, D. Bresson, I. Kessler, M. L. Perrin, A. Guillaudeau, K.
experiments. Durand, S. Ponsonnard, C. Couquet, C. Yardin and C. Mounayer,
"Histopathologic evaluation of arterial wall response to 5
V. LIMITATIONS neurovascular mechanical thrombectomy devices in a swine model,"
AJNR Am. J. Neuroradiol., vol. 34, pp. 2192-2198, Nov-Dec, 2013.
Current studying aims to prove the concept of using RF in [11] M. J. Gounis, A. K. Wakhloo and J. Y. Chueh, "Preclinical
thrombectomy. The principal aim is to examine the investigations for thrombectomy devices--does it translate to
humans?" Stroke, vol. 44, pp. S7-S10, Jun, 2013.
feasibility of the approach, and Tygon tubes occlusion model [12] I. Yuki, I. Kan, A. Golshan, J. Sohn, Y. Murayama, H. V. Vinters and
were used instead of actual blood vessels. Actual blood F. Vinuela, "A swine model to analyze arterial structural changes
vessels are more compliant and may present additional induced by mechanical thrombectomy," AJNR Am. J. Neuroradiol.,
vol. 34, pp. E87-90, Aug, 2013.
complications during retrieval which affects retrieval [13] J. Gralla, G. Schroth, L. Remonda, K. Nedeltchev, J. Slotboom and C.
success. Furthermore, extended benchmarking with Brekenfeld, "Mechanical thrombectomy for acute ischemic stroke:
traditional devices is on-going and will be reported in a thrombus-device interaction, efficiency, and complications in vivo,"
Stroke, vol. 37, pp. 3019-3024, Dec, 2006.
future report. [14] A. I. Brill, "Bipolar electrosurgery: convention and innovation," Clin.
Obstet. Gynecol., vol. 51, pp. 153-158, Mar, 2008.
VI. CONCLUSION [15] N. N. Massarweh, N. Cosgriff and D. P. Slakey, "Electrosurgery:
History, Principles, and Current and Future Uses," J. Am. Coll. Surg.,
A new low force RF-mechanical thrombectomy device is vol. 202, pp. 520-530, 3, 2006.
developed and tested. Clots became strongly adhered to the [16] R. Agah, J. A. Pearce, A. J. Welch and M. Motamedi, "Rate process
device after 3s of RF application, and the clots can be model for arterial tissue thermal damage: Implications on vessel
photocoagulation," Lasers Surg. Med., vol. 15, pp. 176-184, 1994.
successfully retrieved without the need for a large stenting
1352